Skip to main content
NCBI home page
Proceedings of the Royal Society B: Biological Sciences logoLink to Proceedings of the Royal Society B: Biological Sciences
. 2020 Dec 16;287(1941):20202229. doi: 10.1098/rspb.2020.2229

Baffling: a condition-dependent alternative mate attraction strategy using self-made tools in tree crickets

Rittik Deb 1,2,†,✉, Sambita Modak 1, Rohini Balakrishnan 1
PMCID: PMC7779502  PMID: 33323074

Abstract

Intense sexual selection in the form of mate choice can facilitate the evolution of different alternative reproductive strategies, which can be condition-dependent. Tree cricket males produce long-distance acoustic signals which are used by conspecific females for mate localization and mate choice. Our study shows that baffling, an acoustic call amplification strategy employed by male tree crickets using self-made tools, is a classic example of a condition-dependent alternative strategy. We show that though most males can baffle, less preferred males, such as smaller and lower-amplitude callers, predominantly use this alternative strategy. Baffling allows these males to increase their call amplitude and advertisement range, which attracts a higher number of females. Baffling also gives these males a mating benefit because females mate for longer durations with them. Our results suggest that the advantage of baffling in terms of gain in the number of sperm cells transferred while mating is primarily limited to less preferred males, thus maintaining the polymorphism of calling strategies in the population. We summarize that baffling is a condition-dependent strategy used by less preferred tree cricket males to obtain mating benefits.

Keywords: sexual selection, mate choice, Oecanthus henryi, bioacoustics, female preference

1. Introduction

Sexual selection is a dominant process that generates significant variation in biological systems. This variation is not, however, limited to diversity between the sexes. Sexual selection operating independently or in tandem with natural selection can also generate significant diversity within a sex. One of the primary examples of within-sex diversity is alternative reproductive strategies and tactics. Taborsky et al. [1] summarized these as the presence of discontinuous phenotypic traits (physiology, morphology and behaviour) that maximize individual fitness without using the dominant strategy. Generally, such strategies become frequent when intense sexual selection creates biased mating and skewed reproductive success in a population. In nature, the strength of sexual selection is often greater on males and has led to the prevalence of such tactics predominantly in this sex [1,2].

In the tree cricket genus Oecanthus, males attract females using acoustic signals: they primarily call from leaf edges, and females respond by localizing the males [3]. Interestingly, there exists an alternate mode of signalling, where the males make a hole at the centre of a leaf and use it as an acoustic baffle (a sound amplifier) [4,5]. Previous work [6] showed that male tree crickets not only manufacture these tools [7] but do so optimally: they choose larger leaves and make holes in optimal leaf positions, maximizing sound amplification. However, to date, it remains unclear what drives the evolution and maintenance of this alternative behavioural strategy.

In this study, using extensive field sampling and laboratory experiments, we show that baffling propensity is higher in smaller and softer males (the term ‘softer’ refers to lower call amplitude). Interestingly, male body size is known to be under strong selection pressure from female mate choice: females mate longer with larger males [8], leading to the transfer of a higher number of their sperm cells compared with smaller males. Using behavioural experiments in the laboratory and simulations reflecting field conditions, we show that baffling is a successful alternative strategy used by less preferred males to gain mating and reproductive benefits. Our results suggest that the advantage of this strategy is mostly limited to less preferred males, making it a classic example of a condition-dependent, alternative strategy.

2. Results

(a). Who are the bafflers?

Our field sampling (electronic supplementary material, figure S1) revealed that baffling is rare (25 bafflers out of 463 calling males sampled (5%)) in the natural condition (see the electronic supplementary material). An earlier study [6] showed that not all males prefer to call from baffles even when given ideal leaf sizes. Our field observations (field site details: electronic supplementary material, figure S1) revealed that males calling from baffles (bafflers) were smaller in body size than non-baffling males (non-bafflers) (Welch t42.65 = 4.09, p = 0.0002) (figure 1a). We also found that, in the field, bafflers, when calling without a baffle, had lower sound pressure levels (SPLs: a measure of call loudness) than non-bafflers (SPL comparison: Mann–Whitney U-test W=389.5, p = 0.003) (figure 1b; also see the electronic supplementary material, figure S2A, B and table S1 for results of control experiments). Our laboratory experiments showed that propensity to baffle was higher on larger leaves (analysis of deviance: χ23 = 8.63, p = 0.03, electronic supplementary material, table S2B) (concordant with the previous study [6]) and for smaller (analysis of deviance: χ22 = 8.08, p = 0.02, electronic supplementary material, table S2B) and softer males (i.e. males with lower SPL; analysis of deviance: χ21 = 10.42, p = 0.001, electronic supplementary material, table S2B; figure 1c,d). Smaller males baffled with a higher propensity, which increased with an increase in leaf size (extra-large leaf: medium and small males versus large males: proportion test χ21 = 5.95, p = 0.02; large leaf: small versus medium males: proportion test χ21 = 5.95, p = 0.02, large versus medium males p = 0.46; figure 1c). For smaller leaves, the propensity to baffle was not influenced by male size (medium leaf: p = 0.91, small leaf: p = 0.94; figure 1c) owing to lack of baffling using these leaf sizes (shown in [6]). Bafflers when calling without baffles had lower SPL calls than non-bafflers, independent of leaf size (small leaf: sample size too low for statistical testing, SPL comparisons: medium leaf: t18.6 = −9.37, p = 1.75 × 10−8, large leaf: W = 15.5, p = 5.75 × 10−9, extra-large leaf: W = 49.5, p = 1.74 × 10−6; figure 1d).

Figure 1.

Figure 1.

Open in a new tab

Relationship between baffling propensity, male size and call sound pressure level (SPL). (a) Box plots showing the body-size distribution of bafflers and non-bafflers observed in the field. (b) Box plots showing the without-baffle call SPL distribution of bafflers (when forced to call from leaf edge), and non-bafflers, in the field. (c) Bar diagram showing baffling propensity as a function of leaf size and male body size. Predominantly small males were the bafflers across leaf-size classes (see the text for details). Seventeen males of each size class were tested on each leaf size. (d) Box plots showing the distributions of call SPLs of non-bafflers (grey boxes) and bafflers when calling without baffles (white boxes) across three different leaf-size classes. No statistical test was performed for small leaf, as the number of bafflers in the small leaf class was only 1. The asterisk indicates a significant difference.

(b). What are the advantages of baffling?

(i). Increase in active acoustic volume

We found that by baffling, males gained an increase in SPL between 8 and 12 dB (mean ± s.d.: 10.2 ± 1.2 dB, Mann–Whitney U-test W = 284, p = 1.7 × 10−6) in the wild (electronic supplementary material, figure S2A). Concordant with previous work [6], we found that baffling from larger leaves resulted in higher SPL gain (electronic supplementary material, figure S2C and table S3). We calculated the shape and volume of 10 free moving males' active acoustic volumes (electronic supplementary material, figure S3A) while calling from both leaf edge and a baffle. The active acoustic volume of a calling male is defined as the volume within which the call SPL is greater or equal to the hearing threshold of the female. The acoustic volume resembled a three-dimensional figure of ‘8’ (electronic supplementary material, figure S3B–D) with the animal at the centre. This shape was well approximated (R2 = 0.99, p < 2.2 × 10−16; electronic supplementary material, figure S4B) by two identical ellipsoids touching each other along the longest radius (b) representing the body axis of the animal (electronic supplementary material, figure S4A). We calculated active acoustic volumes of males calling at different SPLs (ranging from 58 to 68 dB) with varying baffling gain (ranging from 8 to 12 dB). Concordant with previous results on Oecanthus quadripunctatus from North America [11], we found that baffling caused substantial increment (2.5–11 times) in the overall active acoustic volume (figure 2a).

Figure 2.

Figure 2.

Open in a new tab

Call SPL, female preference and baffling. (a) Comparison between the active acoustic volume of an individual when calling without a baffle (non-baffler volume) and when it is baffling with either 8, 10 or 12 dB increment in call SPL across a range of call SPLs (58–68.5 dB). Inset showing wire-diagram and approximate shape of the active acoustic volume. (b) Female response in the two-speaker choice experiment. Females showed a preference for the louder call (black bars) in all three treatments. (c) Difference in the proportion of females attracted by a single male (same individual) when it was baffling (black bar) versus when not-baffling (grey bar) across 24 simulated natural choruses. (d) Relationship of call SPL (raw amplitude values in micro-pascals centred and scaled to avoid overdispersion) and proportion of females obtained by the caller (pooled across all the animals in 24 choruses for 100 iterations in each chorus) when one individual in each chorus was transformed into a baffler (scenario 2). White circles represent the bafflers. The line shows the fit of the GLMM along with 95% confidence interval (the grey area around the best-fit line). (Online version in colour.)

(ii). Female phonotactic preference for louder (higher amplitude) calls

Using the natural male chorus structures [9], we calculated the SPL differences between male calls that females predominantly encounter in the field (50 dB median, with differences of 3–9 dB; see the electronic supplementary material, figure S5A). Our two-choice phonotactic assay (electronic supplementary material, figure S5B) revealed that females preferred to approach louder calls, even when the difference in SPL was just 3 dB (50 : 53 dB: χ21 = 9, adj. p = 0.003; 50–56 dB: χ21 = 9.8, adj. p = 0.002; 50–59 dB: χ21 = 21 adj. p = 4.59 × 10−6; figure 2b).

(iii). Increase in the proportion of females attracted

We combined the results of active space volume overlap in the field and female preference for louder calls to simulate 24 natural choruses [9] to examine whether bafflers attracted a greater proportion of females. In the choruses where none of the males were baffling (scenario 1, see Material and methods), males with higher amplitude calls obtained a higher proportion of females (electronic supplementary material, figure S6A; generalized linear mixed model (GLMM) with binomial error: fixed effect: scaled call amplitude: 0.4, s.e. = 0.007, z = 52.19, p = 2 × 10−16; random effect: chorus identity (ID): variance 0.2, s.d. 0.45, no. of obs.: 229, group: 24). Next, we converted a randomly chosen male in each chorus into a baffler and found that the proportion of females attracted increased significantly during baffling (figure 2c) in each chorus (pairwise permutation test: baffling versus non-baffling = 4.643, adj. p = 3.43 × 10−6). We found that the proportion of females attracted by the bafflers increased with increasing SPL (figure 2d, scaled call amplitude: 0.54, s.e. = 0.005, z = 104.95, p = 2 × 10−16, random effect: chorus ID: variance 0.14, s.d. 0.38, no. of obs.: 229, group: 24, see the electronic supplementary material, figure S6B for Z-score plot). Baffling allowed males to obtain the highest proportion of females in 19 out of 24 choruses (figure 2d).

(iv). Does baffling offer mating benefits?

It is known that in the absence of acoustic cues, tree cricket females prefer larger males during mating by retaining the spermatophores of the preferred males significantly longer [8], thus allowing longer duration for sperm transfer [10]. Hence, we examined whether females could differentiate between bafflers (smaller/softer call males) and genuine louder/bigger callers during mating. A playback and mating experiment was designed to replicate a natural phonotaxis followed by mating scenario. For the experiment, we chose a total of 20 soft callers (less than 61 dB SPL) and increased their call SPL by 8–12 dB via playback, transforming them into bafflers (SL: soft transformed to loud). We kept the call SPLs of another 20 soft males unaltered (SS: soft remained soft). We performed the same treatment for naturally loud callers (greater than 66 dB SPL) (10 males, LL: loud remained loud; 10 males, LEL: loud transformed to extra loud). If females could identify the genuine loud callers, we expected that they would discriminate between LL and SL males and would not discriminate between SL and SS males. Whereas, if baffling influenced female choice during mating, we expected females to mate longer with the SL males when compared with SS males, and to not discriminate between SL and LL males.

We found that female phonotactic response varied depending on the playback treatment (F3,56 = 47.5, p = 2.06 × 10−15). Females did not differentiate between a true loud caller (LL) and a baffler (SL) in the time taken to approach the call during phonotaxis (SL versus LL, estimate: 7.85, t = 0.56, p = 0.9; figure 3a). However, they showed a faster response towards louder calls (SS versus SL, estimate: 38.5, t = 2.74, p = 0.04; SS versus LL, estimate: 46.35, t = 4.04, p < 0.001; figure 3a). We found that the females corrected their path multiple times and took significantly longer to reach the LEL calls (LL versus LEL: estimate −160.35, t = −11.4, p < 0.01; SL versus LEL: estimate −152.5, t = −9.4, p < 0.01; SS versus LEL: estimate −114, t = −8.1, p < 0.01).

Figure 3.

Figure 3.

Open in a new tab

The benefits of baffling. (a) Phonotaxis duration across four playback treatments. Abbreviations: SS (grey box), soft caller whose call was played back as a soft call; SL (white box), soft caller whose call was played back as a loud call, simulating baffling; LL (right-angled striped box), loud caller whose call was played back as a loud call; LEL (left-angled striped box), loud caller whose call was played back as an extra loud call, simulating baffling. The asterisk indicates a significant difference. (b) Distributions of SPAD across different treatments. These treatments were carried out across different body-sized males: small, medium and large. (c) Number of females attracted by each type of male (loud (greater than 66.1 dB) and soft (less than 61 dB)) when they were baffling (grey boxes) versus when they were not baffling (white boxes). Louder males attracted approximately three females even without baffling (optimum number of females they could mate with within a night), whereas the softer males gained significantly when they baffled. (d) Change in the number of sperm cells transferred over the lifetime (LST across 100 nights) when loud and large, soft and large, loud and small, and soft and small males were baffling versus when they were not baffling. The large males were restricted to a maximum of 2 matings night−1, whereas the small males were restricted to a maximum of 5 matings night−1 (see Material and methods). (Online version in colour.)

We found that spermatophore attachment duration (SPAD) varied depending on both playback treatment and male body size (playback treatment: F3,54 = 42.02, p = 3.79 × 10−14; male size class: F2,54 = 181.88, p = 2 × 10−16) (figure 3b). SPAD was shorter for softer (SS) males in all size classes (LEL versus SS: 695.8, t = 5.95, p = 2.03 × 10−7; LL versus SS: 743.7, t = 6.551, p = 2.21 × 10−8; SL versus SS: 637.23, t = 6.99, p = 4.28 × 10−9; figure 3b), indicating female discrimination against soft callers during mating. SPAD was comparable between SL, LL and LEL males (LEL versus SL: −58.5, t = −0.49, p = 0.62; LEL versus LL: 47.9, t = 0.35, p = 0.72; figure 3b), indicating the efficacy of baffling in inducing females into mating for longer durations with the less preferred softer males than they otherwise would have. Concordant with the previous study [8], SPAD increased with male body size (large versus small: −1589.8, t = −19.07, p = 2 × 10−16; large versus medium: −937.5, t = −7.79, p = 2.12 × 10−10; figure 3b). However, it was evident that the differences in SPAD between larger and smaller males declined when the smaller males baffled (figure 3b). These results mean that a baffler can induce a female into approaching and mating with it for longer durations than if it did not baffle.

(v). Final puzzle: why don't larger and louder males baffle as much?

We hypothesized that as louder and larger males attracted a significant number of mates per night (median 2.6) and had very high SPAD (approx. 45 min female−1) [8] even without baffling, their gain by baffling would be minimal, making baffling inconsequential. We compared the gain in SPAD by baffling for soft males (i.e. SL versus SS) across two body-size classes (large and small; figure 3b) and found that small males (median gain: 624 s) gained significantly longer SPAD than large males (median gain: 483.5 s) by baffling (W = 4252.5, p = 0.04).

We used simulations to compare the advantages of baffling (numbers of females attracted and sperm transferred) between different classes of males in natural choruses. Our simulation used a nested design. In each chorus, we randomly chose a loud male and examined the proportion of females it attracted when it was baffling and when not baffling. For each of these loud baffling males, we considered two scenarios: that it was a (i) large and (ii) small male and estimated the number of sperm transferred during mating (considering remating within a night). We simulated each chorus for 100 nights to calculate a male's lifetime sperm transfer (LST) success. We calculated the difference of LST for a loud, large male when it was baffling and when it was not baffling, and repeated the same analysis for loud, small males. We repeated this analysis for soft, large and soft, small males and compared the distributions to understand the gain differences.

We found that soft non-baffler males, soft_baffler males, loud non-baffler males and loud_baffler males obtained a median of 0.35, 2.5, 2.6 and 7.7 females night−1, respectively (figure 3c). Our analysis showed that it was only advantageous for a soft male to baffle as the louder males were attracting close to the maximum number of females they could mate with within a night (around three, calculated as an activity period divided by mating duration) even without baffling (figure 3c).

We found that gain in LST when baffling was significantly different between the different groups of males (figure 3d; electronic supplementary material, figure S8, table S4a). The median gain in LST was highest for soft, large males (3.5 × 106) followed by soft, small males (2.3 × 106) and loud, small males (2.2 × 106) (figure 3d and the electronic supplementary material, table S4a,b). The gain for the loud, large males was three orders of magnitude less compared with other males and was negligible (3 × 103). We found that the LST gain obtained by soft, large males was more variable (figure 3d). These results support our hypothesis regarding a differential gain between preferred (larger and louder) and less preferred males through baffling.

3. Discussion

(a). Baffling: a classic case of conditional alternate tactics

Earlier studies [4,6,7,11] had described baffling behaviour, optimization and the underlying physical principle in detail. However, none of them examined the evolutionary significance of this unique behaviour. In this study, we show that the less preferred males (smaller and with lower-amplitude calls) were more likely to use this alternative signalling tactic to gain both mating and sperm transfer benefits. The use of this alternative strategy should allow these males to reduce the inherent reproductive bias present in the species [8]. Such alternative strategies based on the male condition have theoretically been classified as condition-dependent strategies, where genetically monomorphic individuals choose to express a particular strategy depending on the status/state/condition of the individual [2,12]. In condition-dependent strategies, the average fitnesses of the alternative strategies are not equal, but provide higher fitness returns to the users (higher than if they were using the other dominant strategy) depending on their particular state [2,12–16].

Gross [12] laid out the critical characteristics that are essential to be examined to classify an alternative strategy as condition-dependent: (i) presence of choice, (ii) genetic monomorphism with respect to the strategy, (iii) condition/status of individuals, (iv) fitness difference between alternatives, and (v) fitness advantage of the strategy. Despite decades of research, only a few studies have examined all these characteristics to classify an alternative tactic as truly condition-dependent ([1,2], as reviewed in [12]). In our study, we addressed all the critical points to show that baffling is a classic case of a condition-dependent alternative strategy. (i) This study, in conjunction with earlier work [6], establishes baffling as a facultative behaviour by showing that most males can baffle given an ideal leaf size, but only a fraction of them effectively used this strategy in the wild. These findings establish that males could choose to use this tactic. (ii) During rearing, we observed that newly emerged adult males kept in isolation also used this tactic (R. Deb 2015, unpublished data). This, coupled with the fact that all males can baffle [6], indicates that all males are inherently capable of expressing this behaviour. (iii) We show that this tactic was predominantly used by smaller males and/or those with lower-amplitude calls, establishing that male status/condition plays an important role in the expression of this tactic. (iv) We provide evidence that the average fitness of baffling (alternative tactics) is higher than the dominant tactic. (v) Finally, our findings indicate that the advantage gained by using this tactic was significant only for less preferred males.

(b). How is the polymorphism maintained?

The maintenance of multiple signalling/mating strategies in a population has been an intriguing problem in evolutionary biology [17,18]. Alternative strategies provide differential gains across different status/condition of individuals and might not render equal benefits to all individuals [1,12,19,20]. Often, the average fitnesses of the alternative strategies are not equal, but provide higher fitness returns to the users (than if they were using the other strategy) depending on their particular state [14,19,21,22]. Such differential benefits have been hypothesized to exist because mating benefits and opportunities provide diminishing returns to any dominant individual/strategy beyond the optimum, owing to costs and lack of resource utilization [23,24]. A dominant male's reproductive success often reaches a plateau because it can mate only with a finite number of females within a given time [19,23].

Concordantly, in this study, we found that though most males that baffled attracted more females, it was likely that the preferred males were unable to mate with all of them. The preferred males were able to attract an optimum number of females even without baffling, rendering baffling gain inconsequential. However, the benefit obtained by less preferred males from baffling was significant as they increased their mating opportunities significantly. We also hypothesized that, as preferred males mated for extended durations (approx. 45 min), their sperm transfer rate would reach a plateau [10,25,26] even without baffling. Hence, any increment in SPAD owing to baffling will only incur minimal benefits to the preferred males. Concordant with these hypotheses, we found that the reproductive gain, measured as LST [10,27–30], was not comparable across different males. Less preferred males gained more by baffling, whereas larger and louder (preferred) males' gain was negligible. We also found that the phonotactic females took significantly longer and made multiple errors while localizing the inherently loud males who were transformed into bafflers (LEL), even in our simple experimental set-up (probably as loudness was close to female auditory saturation), which should further reduce/delay female visitation. We suggest that baffling behaviour coexists with other calling strategies as its benefits are coupled with the male status/condition.

Apart from these condition-dependent differential costs and benefits, bafflers might also face generic costs such as searching for ideal leaves, manufacturing cost, predation owing to higher call SPL and lesser vigilance capability. These generic costs should decrease the advantage of baffling and can potentially inhibit a male in a favourable reproductive condition from baffling.

4. Material and methods

(a). Who are the bafflers: preferred or non-preferred males?

(i). Examining body size and call sound pressure level of baffling males in the field

Measuring body size and sound pressure level. We collected 38 non-bafflers and 25 bafflers from field sites and measured their body lengths (see the electronic supplementary material for details). The body lengths of bafflers were compared with non-bafflers using Welch's t-test.

We measured the SPL of 17 baffling and 30 non-baffling males using a Bruel and Kjaer SPL Meter (Bruel & Kjaer, Naerum, Denmark) (see the electronic supplementary material for details). Post SPL measurement, the baffling animal was disturbed such that it moved to a new leaf. After the animal started calling from its new position without a baffle, we again measured its call SPL three times, 10, 60 and 120 min post-disturbance to evaluate whether disturbance caused a decline in SPL (using paired t-tests or Wilcoxon-signed rank test). Similarly, for an independent set of non-baffling males (n = 20), we measured call SPL, disturbed the animal and re-recorded call SPLs at 10, 60 and 120 min post-disturbance. We compared pre- and post-disturbance SPLs (call amplitudes) using paired t-tests or Mann–Whitney U-test. All the SPLs measured in a dB scale were converted to linear raw amplitude for statistical analysis using the following formula:

A2=A1⋅10(GdB/20),

where A1 is the reference amplitude, 20 µPa (universal standard), GdB is the SPL measured in dB using the SPL meter and A2 is the calculated amplitude.

The call SPL of bafflers when calling without a baffle was compared (as SPL had high within night repeatability [8]) with that of the non-bafflers using a Welch's t-test. The call SPL of bafflers when calling with and without a baffle was compared using a paired t-test or Mann–Whitney U-test.

(ii). Effect of male size and call sound pressure level on baffling probability across different leaf-size classes

We conducted this experiment in the laboratory on 51 males in complete darkness during the peak calling period of Oecanthus henryi males (19.00 and 21.30). The animals were chosen from three body-size classes (small (less than 10.9 mm), medium (10.91–12.1 mm), and large (greater than 12.21 mm)), each class equally represented (n = 17) (based on population body-size distribution [8]) (see the electronic supplementary material for details).

The set-ups consisted of a small Hyptis suaveolens (host plant) twig (15 cm in length) embedded in a thermocol piece and covered by a plastic jar. Each animal was released on four different leaf sizes (randomized order for each animal) over four consecutive nights and observed during the entire calling period. The leaf sizes were small (3.5 (±0.2) × 2.5 (±0.11) cm), medium (4.5 (±0.2) × 3.5 (±0.1) cm), large (6.5 (±0.2) × 5 (±0.1) cm) and extra-large (11 (±0.4) × 9 (±0.2) cm) (classes were defined based on leaf-size distribution [6]). We measured the SPL of the calling animals with and without baffling after lifting off the plastic jar.

We used a GLMM with the binomial error structure (R-package lme4, [31]) to examine if baffling propensity was influenced by body size, without-baffle call SPL, and leaf size, with animal ID as a random effect (see the electronic supplementary material). We converted the SPL into two categorical variables: soft (less than 63.8 dB) and loud (greater than 63.8 dB) (population SPL distribution: 63.8 ± 2.55 dB) [8]. We divided the animals into size classes and compared their baffling probability using proportion tests. The distributions of call SPL without baffling were compared between bafflers and non-bafflers using Welch's t-test or Mann–Whitney U-test.

(b). Advantages of baffling

(i). Experiment: effect of male body size and leaf size on the baffling gain

We examined whether the gain in SPL by baffling was dependent on (i) the baffling leaf size and (ii) the baffler's body size. We experimented on 30 males (10 large, 10 medium and 10 small males). We prepared four experimental set-ups (identical to the earlier design). First, we released the animal on a set-up consisting of a medium leaf (with no baffle hole) and measured the call SPL (without baffle). Next, on the same night, we moved the animal to three other set-ups, which consisted of small, medium and large leafs, respectively, with an artificially made baffle hole (see the electronic supplementary material). The order of presentation of leaves with baffling holes was randomized for each animal. We calculated the gain in SPL by baffling by deducting the SPL of the animal when calling without baffle from the SPL measured while the animal was calling from baffles in small, medium and large leaves. We examined the effect of leaf size and male size on the gain in SPL via baffling using a GLMM with an animal ID number as the random effect.

(ii). Increase in active acoustic volume

We measured the active acoustic volume of a caller when calling with and without baffle (see the electronic supplementary material, figures S3 and S4 for details, using a set-up inspired by Forrest et al. [11]) across several non-baffling call SPLs, ranging from 58 to 68.5 dB (increment of 0.5 dB), and calculated non-baffling active acoustic volume. We calculated the baffling volume by increasing each non-baffling SPL by 8, 10 and 12 dB SPL and compared the gain in active acoustic volume through baffling.

(iii). Effect of baffling on female phonotactic preference

We chose three combinations of SPL pairs for playback: 50–53 dB, 50–56 dB and 50–59 dB, with a difference of 3, 6 and 9 dB, respectively, and examined female preference for SPL (see the electronic supplementary material). We randomized the directionality of the louder speaker for each test trial for each animal. We tested 16 (50–53 dB), 20 (50–56 dB) and 21 (50–59 dB) females for the three treatments. We performed a χ2 test to examine if the females preferred the louder calls in each trial while controlling for false discovery rate.

(iv). Effect of baffling on male-mating opportunities

For the simulation, 24 natural chorus spatial maps along with measured male call SPLs were used [9]. To measure the gain in mating opportunities via baffling, we simulated each chorus twice, scenario 1: without any baffler in the chorus and scenario 2: with a randomly chosen male transformed into a baffler by increasing its call SPL (see the electronic supplementary material). For the simulation, we considered only the females that occurred inside the active acoustic volume of a calling male (see the electronic supplementary material). We kept the calling male to female sex ratio for each chorus equal (1 : 1). If a female was found inside the active acoustic volume of multiple calling males, it was assigned to the louder caller (see the electronic supplementary material) or a randomly chosen male if the SPL values were comparable (see the electronic supplementary material). At the end of the simulation run, we recorded the number of females obtained by each male. This process was iterated 100 times for each chorus, representing 100 nights for each male, to calculate an estimate of the lifetime mating success for each male (see the electronic supplementary material). At the end of 100 iterations, the proportion of females obtained by a baffler male while baffling and not baffling was compared across all the 24 choruses using the pairwise-randomization test. We performed a GLMM with binomial error, where the proportion of females obtained was the response variable, the call amplitude (centred and scaled in R to avoid overdispersion) without baffle was the predictor and chorus ID was the random variable across both the simulation scenarios. We calculated Z-scores for all the males based on the proportion of females attracted and plotted them against the call SPLs (centred and scaled raw amplitude).

(v). Effect of baffling on female phonotaxis and mating behaviour

We examined (i) whether females preferred bafflers (louder males) both during phonotaxis (i.e. faster response) and mating (i.e. longer SPAD), and (ii) if the females could differentiate between a naturally loud male and an artificially loud (baffling) male (see the electronic supplementary material). We had two treatment effects: (i) body size of male (small, medium or large) and (ii) playback treatment (SS, SL, LL and LEL) resulting in 12 treatment combinations (figure 3). We divided the whole sample subset into 12 combinations, such as S_SS (small and soft male, call remained soft), S_SL (small and soft male, call transformed to a baffler), S_LL (small and loud male remained loud) and S_LEL (small and loud male, call transformed to a baffler), M_SS, M_SL, M_LL, M_LEL (for medium-sized males) and L_SS, L_SL, L_LL, L_LEL (for large-sized males).

We calculated the latency of response by measuring the time taken by each female to reach the playback speaker (see the electronic supplementary material) and compared them using ANOVA and post hoc Tukey's tests. We measured SPAD as a proxy for female-mating preference for a male [8] and examined the effect of body size and playback treatment on the SPAD using ANOVA.

(vi). Why don't larger and louder males baffle as much as smaller and softer males?

Is the gain in SPAD owing to baffling comparable between small_soft and large_soft males? We calculated the gain in SPAD for small_soft males when they were transformed into bafflers (i.e. S_SL–S_SS) using the result of the previous experiment (figure 3). Similarly, we calculated the gain in SPAD for large_soft males when they were transformed into bafflers (i.e. L_SL–L_SS). As our earlier experiment was performed on independent individual animals, we calculated all possible differences in SPAD between each S_SL and each S_SS data points and generated the distribution of the differences for small males. Similarly, we also prepared the distribution of the difference between L_SL and L_SS treatments. We compared the distributions using the Mann–Whitney U-test.

Is the gain of baffling (number of females attracted and amount of sperm transferred) comparable between preferred (large and loud) and less preferred (small and soft) males? We simulated each of our 23 natural choruses four times. In the first simulation, in each chorus, we randomly chose a loud male (SPL > 66.1 dB) and calculated the number and proportion of females it could attract over its lifetime (i.e. 100 nights). In the second simulation, we transformed these loud males into bafflers (without altering their position in the chorus) and recalculated the number and proportion of females they attracted over their lifetime (see the electronic supplementary material). In the third round of the simulation, we chose a soft male in each chorus and examined the number and proportion of females they attracted over their lifetime. In our fourth round of the simulation, we transformed these soft males into bafflers and re-examined the number and proportion of females they attracted over their lifetime. We divided the number of females attracted over their lifetime by 100 (i.e. 100 nights of calling) to calculate an average number of females these males attracted per night by calling with or without baffle. We plotted these values for loud_baffler, loud_non_baffler, soft_baffler and soft_non_baffler across 23 choruses and compared the medians.

We calculated the sperm transfer function, i.e. the number of sperm cells transferred as a function of time, for three body-size classes of males using the data from a conspecific species Oecanthus nigricornis [10] (electronic supplementary material, table S4 and figure S7; see the electronic supplementary material). We multiplied these sperm transfer functions with the respective mating durations (figure 3b) to calculate the amount of sperm that a male could transfer for a given mating depending on its call SPL and body size. Next, we multiplied these with number of mates a male can mate with, in its lifetime (i.e. 100 nights) across each chorus. We calculated different remating scenarios per night, where a male could mate with 1–5 mates night−1. We limited the maximum number of matings for a large male to 2 mates per night−1 (approx. 84 min mating; figure 3b) and for a small male to 5 mates per night−1 (approx. 75 min mating; figure 3b). We deducted the LST of a male when it was baffling from the LST when it was not baffling to calculate the gain in sperm transfer by baffling (figure 3d; electronic supplementary material, S8).

Supplementary Material

Supplementary information
rspb20202229supp1.docx (1.3MB, docx)
Reviewer comments
rspb20202229_review_history.pdf (747.6KB, pdf)

Supplementary Material

Supplementary code and raw data
rspb20202229supp2.zip (197.7KB, zip)

Acknowledgements

We thank the associate editor, board member and the two anonymous reviewers for encouraging and critical comments. We thank R. Manjunatha, N. Ashoka, U.N. Narsimhamurthy, C.N. Chandra and K. V. Rao for help with the collection of animals and call recordings in the field. We also thank D. Nandi, M. Bhattacharya and S. Pulla for helpful discussions regarding the analysis and coding.

Data accessibility

The codes for measuring dipole activity acoustic volume, estimating the proportion of females attracted by each male in a chorus, and the raw data are available as part of the electronic supplementary material.

Authors' contributions

R.D.: conceptualization, methodology, software, validation, formal analysis, investigation, resources, data curation, writing—original and final draft, review and editing, visualization, supervision, project administration and funding acquisition. S.M.: formal analysis, investigation, resources, data curation and critical manuscript review. R.B.: conceptualization, resources, writing—manuscript editing and critical review, supervision, project administration and funding acquisition.

Competing interests

The authors declare no competing interests (financial or non-financial).

Funding

Fieldwork was supported by the Ministry of Environment, Forests and Climate Change, Govt. of India and consumables by the DBT-IISc Partnership Program, Govt. of India. Equipment used was funded by DST-FIST (Fund for Improvement of Science and Technology Infrastructure, Govt. of India). R.D. was supported by the Council of Scientific and Industrial Research 09/079(2199)/2008-EMR-I., Govt. of India. The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

References

  • 1.Taborsky M, Oliveira RF, Brockmann HJ. 2008. The evolution of alternative reproductive tactics: concepts and questions. In Alternative reproductive tactics: an integrative approach (eds Oliveira RF, Taborsky M, Brockmann HJ), pp. 1–22. Cambridge, UK: Cambridge University Press. [Google Scholar]
  • 2.Shuster SM. 2010. Alternative mating strategies. In Evolutionary behavioral ecology (eds DF Westneat, CW Fox), pp. 434–450. New York, NY: Oxford University Press. [Google Scholar]
  • 3.Walker TJ. 1957. Specificity in the response of female tree crickets (Orthoptera, Gryllidae, Oecanthinae) to calling songs of the males. Ann. Entomol. Soc. Am. 50, 626–636. [Google Scholar]
  • 4.Prozesky-Schulze L, Prozesky OPM, Anderson F, Van Der Merwe GJJ. 1975. Use of a self-made sound baffle by a tree cricket. Nature 255, 142–143. ( 10.1038/255142a0) [DOI] [Google Scholar]
  • 5.Forrest TG. 1982. Acoustic communication and baffling behaviors of crickets. Florida Entomol. 65, 33–44. [Google Scholar]
  • 6.Mhatre N, Malkin R, Deb R, Balakrishnan R, Robert D. 2017. Tree crickets optimise the acoustics of baffles to exaggerate their mate-attraction signal. eLife 6, e2763 ( 10.7554/eLife.32763) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Mhatre N. 2018. Tree cricket baffles are manufactured tools. Ethology 124, 691–693. ( 10.1111/eth.12773) [DOI] [Google Scholar]
  • 8.Deb R, Bhattacharya M, Balakrishnan R. 2012. Females of a tree cricket prefer larger males but not the lower frequency male calls that indicate large body size. Anim. Behav. 84, 137–149. ( 10.1016/j.anbehav.2012.04.020) [DOI] [Google Scholar]
  • 9.Deb R, Balakrishnan R. 2014. The opportunity for sampling: the ecological context of female mate choice. Behav. Ecol. 25, 967–974. ( 10.1093/beheco/aru072) [DOI] [Google Scholar]
  • 10.Brown WD. 1997. Courtship feeding in tree crickets increases insemination and female reproductive life span. Anim. Behav. 54, 1369–1382. ( 10.1006/anbe.1997.0541) [DOI] [PubMed] [Google Scholar]
  • 11.Forrest TG. 1991. Power output and efficiency of sound production by crickets. Behav. Ecol. 2, 327–338. ( 10.1093/beheco/2.4.327) [DOI] [Google Scholar]
  • 12.Gross MR. 1996. Alternative reproductive strategies and tactics: diversity within sexes. Trends Ecol. Evol. (Amst.) 11, 92–98. ( 10.1016/0169-5347(96)81050-0) [DOI] [PubMed] [Google Scholar]
  • 13.Gross MR. 1984. Sunfish, salmon, and the evolution of alternative reproductive strategies and tactics in fishes. In Fish reproduction: strategies and tactics (eds GW Potts, RJ Wooton), pp. 55–73. London, UK: Academic Press. [Google Scholar]
  • 14.Badyaev AV. 2002. Paternal care as a conditional strategy: distinct reproductive tactics associated with elaboration of plumage ornamentation in the house finch. Behav. Ecol. 13, 591–597. ( 10.1093/beheco/13.5.591) [DOI] [Google Scholar]
  • 15.Studd MV, Robertson RJ. 1985. Sexual selection and variation in reproductive strategy in male yellow warblers (Dendroica petechia). Behav. Ecol. Sociobiol. 17, 101–109. [Google Scholar]
  • 16.Gerhardt HC, Huber F. 2002. Acoustic communication in insects and anurans: common problems and diverse solutions. Chicago, IL: University of Chicago Press. [Google Scholar]
  • 17.Gadgil M, Taylor CE. 1975. Plausible models of sexual selection and polymorphism. Am. Nat. 109, 470–472. [Google Scholar]
  • 18.Gadgil M. 1972. Male dimorphism as a consequence of sexual selection. Am. Nat. 106, 574–580. [Google Scholar]
  • 19.Shuster SM, Wade MJ. 2003. Mating systems and strategies. Princeton, NJ: Princeton University Press. [Google Scholar]
  • 20.Taborsky M, Brockmann HJ, Kappeler P. 2010. Alternative reproductive tactics and life history phenotypes. In Animal behaviour: evolution and mechanisms (ed. P. Kappeler), pp. 537–586. Berlin, Germany: Springer. [Google Scholar]
  • 21.Reynolds JD, Gross MR. 1990. Costs and benefits of female mate choice: is there a lek paradox? Am. Nat. 136, 230–243. [Google Scholar]
  • 22.Bailey WJ, Field G. 2000. Acoustic satellite behaviour in the Australian bushcricket Elephantodeta nobilis (Phaneropterinae, Tettigoniidae, Orthoptera). Anim. Behav. 59, 361–369. ( 10.1006/anbe.1999.1325) [DOI] [PubMed] [Google Scholar]
  • 23.Waltz EC. 1982. Alternative mating tactics and the law of diminishing returns: the satellite threshold model. Behav. Ecol. Sociobiol. 10, 75–83. [Google Scholar]
  • 24.Real LA. 1980. On uncertainty and the law of diminishing returns in evolution and behavior. In Limits to action: the allocation of individual behavior (ed. JER Staddon), pp. 37–64. New York, NY: Academic Press. [Google Scholar]
  • 25.Sakaluk SK. 1984. Male crickets feed females to ensure complete sperm transfer. Science 223, 609–610. ( 10.1126/science.223.4636.609) [DOI] [PubMed] [Google Scholar]
  • 26.Simmons LW, Wernham J, García-González F, Kamien D. 2003. Variation in paternity in the field cricket Teleogryllus oceanicus: no detectable influence of sperm numbers or sperm length. Behav. Ecol. 14, 539–545. [Google Scholar]
  • 27.Oberhauser KS. 1989. Effects of spermatophores on male and female monarch butterfly reproductive success. Behav. Ecol. Sociobiol. 25, 237–246. ( 10.1007/BF00300049) [DOI] [Google Scholar]
  • 28.Simmons LW. 1988. Male size, mating potential and lifetime reproductive success in the field cricket, Gryllus bimaculatus (De Geer). Anim. Behav. 36, 372–379. ( 10.1016/S0003-3472(88)80008-3) [DOI] [Google Scholar]
  • 29.Brown WD. 1999. Mate choice in tree crickets and their kin. Annu. Rev. Entomol. 44, 371–396. [DOI] [PubMed] [Google Scholar]
  • 30.Brown WD. 2008. Size-biased mating in both sexes of the black-horned tree cricket, Oecanthus nigricornis Walker (Orthoptera: Gryllidae: Oecanthinae). J. Insect Behav. 21, 130–142. ( 10.1007/s10905-007-9112-1) [DOI] [Google Scholar]
  • 31.Darwin C. 1871. The descent of man: and selection in relation to sex, 2nd edn. London, UK: John Murray. [Google Scholar]
  • 32.Andersson M, Iwasa Y. 1996. Sexual selection. Trends Ecol. Evol. 11, 53–58. ( 10.1016/0169-5347(96)81042-1) [DOI] [PubMed] [Google Scholar]
  • 33.Bateson P. (ed.) 1983. Mate choice, 1st edn Cambridge, UK: Cambridge University Press. [Google Scholar]
  • 34.Rowe L, Houle D. 1996. The lek paradox and the capture of genetic variance by condition dependent traits. Proc. R. Soc. Lond. B 263, 1415–1421. ( 10.1098/rspb.1996.0207) [DOI] [Google Scholar]
  • 35.Kotiaho JS. 2001. Costs of sexual traits: a mismatch between theoretical considerations and empirical evidence. Biol. Rev. Camb. Philos. Soc. 76, 365–376. ( 10.1017/s1464793101005711) [DOI] [PubMed] [Google Scholar]
  • 36.Grafen A. 1990. Biological signals as handicaps. J. Theor. Biol. 144, 517–546. ( 10.1016/s0022-5193(05)80088-8) [DOI] [PubMed] [Google Scholar]
  • 37.Ghoul M, Griffin AS, West SA. 2014. Toward an evolutionary definition of cheating. Evolution 68, 318–331. ( 10.1111/evo.12266) [DOI] [PubMed] [Google Scholar]
  • 38.Bates D, Maechler M, Bolker B, Walker S. 2015. Fitting linear mixed-effects models unsing lme4. J. Stat. Soft. 67, 1–48. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary information
rspb20202229supp1.docx (1.3MB, docx)
Reviewer comments
rspb20202229_review_history.pdf (747.6KB, pdf)
Supplementary code and raw data
rspb20202229supp2.zip (197.7KB, zip)

Data Availability Statement

The codes for measuring dipole activity acoustic volume, estimating the proportion of females attracted by each male in a chorus, and the raw data are available as part of the electronic supplementary material.

Articles from Proceedings of the Royal Society B: Biological Sciences are provided here courtesy of The Royal Society

RESOURCES